Binding Kinetics versus Affinities in BRD4 Inhibition - Journal of

Aug 11, 2015 - Arrowsmith , C. H.; Bountra , C.; Fish , P. V.; Lee , K.; Schapira , M. Epigenetic ...... Machine-Learning-Assisted Approach for Discov...
0 downloads 0 Views 2MB Size
Subscriber access provided by Stockholm University Library

Article

Binding Kinetics versus Affinities in BRD4 Inhibition Ming Kuang, Jingwei Zhou, Laiyou Wang, Zhihong Liu, Jiao Guo, and Ruibo Wu J. Chem. Inf. Model., Just Accepted Manuscript • DOI: 10.1021/acs.jcim.5b00265 • Publication Date (Web): 11 Aug 2015 Downloaded from http://pubs.acs.org on August 15, 2015

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Journal of Chemical Information and Modeling is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Chemical Information and Modeling

Binding Kinetics versus Affinities in BRD4 Inhibition Ming Kuang1,2, Jingwei Zhou2, Laiyou Wang1, Zhihong Liu2, Jiao Guo1,*, and Ruibo Wu2,*

1

Guangdong Metabolic Diseases Research Center of Integrated Chinese and Western

Medicine, Guangdong TCM Key Laboratory against Metabolic Diseases,Institute of Chinese Medical Sciences, Guangdong Pharmaceutical University, Guangzhou 510006, P.R. China 2

School of Pharmaceutical Sciences, Sun Yat-sen University, Guangzhou 510006, P.R. China

*To whom correspondence should be addressed. E-mail: [email protected] ; [email protected]

TOC

ACS Paragon Plus Environment

Journal of Chemical Information and Modeling

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Abstract: Bromodomains (BRDs) are protein modules that selectively recognize histones as a “reader” by binding to acetylated lysine substrate. The human BRD4 has emerged as a promising drug target for a number of disease pathways and several potent BRD inhibitors are discovered experimentally recently. However, the detailed inhibition mechanism especially for the inhibitor binding kinetics is not clear. Herein, by employing classical molecular dynamics (MD) and state-of-the-art density functional QM/MM MD simulations, the dynamic characteristics of ZA-loop in BRD4 are revealed. And then the correlation between binding pocket size and ZA-loop motion are elucidated. Moreover, our simulations found that the compound (-)-JQ1 could be accommodated reasonable in thermodynamics whereas it is infeasible in binding kinetics against BRD4. Its racemate (+)-JQ1 is proved to be both thermodynamically reasonable and kinetics achievable against BRD4, which could explain the previous experimental results that (+)-JQ1 shows high inhibitory effect towards BRD4 (IC50 is 77 nM) while (-)-JQ1 is inactive (> 10 µM). Furthermore, the L92/L94/Y97 in ZA-loop and Asn140 in BC-loop are identified to be critical residues in (+)-JQ1 binding/releasing kinetics. All these findings shed lights on further selective inhibitor design towards BRD family, by exploiting the non-negligible ligand binding kinetics features and flexible ZA-loop motions of BRD, instead of only the static ligand-protein binding affinity.

Keywords: binding kinetics; loop motion; BRD4; molecular dynamics; QM/MM;

ACS Paragon Plus Environment

Page 2 of 23

Page 3 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Chemical Information and Modeling

Introduction Epigenetic regulation of gene expression is currently the focus of intensive research in the postgenomic era.1-5 Histones are subjected to a range of post-translational modifications.1, 6 The histone modifications, usually namely ‘histone code’,7 are found mainly on the N-terminal histone tails that protrude from the globular core of the nucleosome and include acetylation, methylation, phosphorylation, ribosylation, biotinylation, citruillination and SUMOylation.1,

8

Each modification can affect

chromatin structure, but the overall state of chromatin is ultimately determined by combinations of these modifications.9 “Readers” of epigenetic marks, which recognize covalent modifications of histone proteins or DNA, are structurally diverse proteins each containing one or more evolutionarily conserved effector modules.10-13 Among of the “Readers”, bromodomains (BRDs), which are protein modules containing approximately 110 amino acids and including a left-handed four-helix bundle (αZ, αA, αB, αC) as well as the varying loop regions (ZA and BC, as shown in Figure 1.), are the only interaction modules that specifically recognize acetylated lysine residues of histones during transcriptional activation.14-18 Many of BRDs are regulators of gene transcription such as histone acetyltransferases, components of chromatin remodeling complexes, and methyltransferases.16,

19

Human proteome

analysis has revealed that there are eight distinct BRD families, representing 61 different BRDs from 46 separate proteins, although others may still be undiscovered.11, 16 The bromodomain and extra terminal (BET, belong to family II BRDs) includes BRD2, BRD3, BRD4 and BRDT, and they exhibits a high level of sequence conservation.20-22 The BRD4 has emerged as promising drug targets for a number of disease pathways that are characterized by changing the epigenetic cell signature.23-25 Recently,

Filippakopoulos

et

al

23

have

identified

a

highly

potent

methyltriazolodiazepine-like inhibitor of BRDs, (+)-JQ1, which has very high affinity with an IC50 value of 77 nM for BRD4, but the (-)-JQ1 stereoisomer is inactive against BRD4 with the IC50 value above 10,000 nM, as shown in Figure 1. Meanwhile,

ACS Paragon Plus Environment

Journal of Chemical Information and Modeling

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

it has been proposed that the plasticity and accessibility of the acetyl-lysine binding site to accommodate diverse inhibitors are likely due to the flexible ZA-loop.23, 26 However, the detailed loop motion and its functional roles are still not clear. Herein, by employing molecular dynamics (MD) simulations on BRD4 with explicit solvent at the molecular mechanics or hybrid quantum mechanical/molecular mechanical (QM/MM) level, we have identified the exact optimal binding modes of the two racemates ((+)-JQ1 and (-)-JQ1) and revealed the flexibility of ZA-loop in BRD4. Meanwhile, we illuminate the relationship between pocket size and ZA-loop motion, which is induced by the ligands. Furthermore, the functional roles of ZA-loop in regulating the binding affinity with JQ1 and mechanism details on the releasing/binding kinetics of (+)-JQ1 are elucidated. All of these findings are guidable for selective inhibitor design towards BET family.

Figure 1. (a) Ribbon illustration of the crystal structure of BRD4 complex with (+)-JQ1 (PDB code 3MXF). The four α-helices and two important loops (ZA-loop & BC-loop, around binding site) are marked. (b,c) Chemical structures and IC50 values of (+)-JQ1 and (-)-JQ1 inhibitors (chiral atoms are highlighted by blue asterisk). The yellow sticks structure indicates the (+)-JQ1 in crystal structure, and the red solid/dashed wedge lines indicate the different orientation of (-)-JQ1 in BRD4. Asn140 is a conserved residue in BRDs and forms strong hydrogen bond with JQ1 ligands.

ACS Paragon Plus Environment

Page 4 of 23

Page 5 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Chemical Information and Modeling

Methods 1. Classical MM MD simulations. The electroneutral enzyme-ligand complex was built from the crystal structure of BRD4 binding with (+)-JQ1 (PDB: 3MXF).23 The ff99SB force field force field

28

27

and GAFF

were employed for the proteins and the two JQ1 ligands respectively.

The partial atomic charges of these small molecule ligands were fitted from the restrained electrostatic potential (RESP) 29 charge at HF/6-31G* theoretical level by using Gaussian09 package.30 The key charge parameters of JQ1 compounds were provided in Table S1, and their surrounding active site environment was represented in Figure S1. The amino acid side chain protonation states were assigned using default AMBER protonation states, and the protonation states of the key residues in BRD4 were also checked manually. His71 was protonated at the epsilon position to form hydrogen bonds with the water molecule nearby. And then the whole system was solvated within a cubic box of TIP3P water molecules

31

by extending 8 Å from the

BRD4 protein surface. Finally, the initial coordinates and topology files were generated by the tleap program in AMBER12.32 Before the heating MD simulation, the system was optimized by three steps. First, the water molecules were minimized whereas keeping the protein and substrates constrained. Then, the side chains of the protein were relaxed while keeping the main chain restrained. Finally, the entire system was optimized. For each minimization step, the conjugate gradient iterations were carried out for 2,000 cycles after performing 2,000 steps steepest descent energy minimization. After the optimization, the system was heated up from 0 K to 310 K gradually under the NVT ensemble for 50 ps. Then, 50 ps MD simulations were performed under the NPT ensemble to relax the system density to about 1.0 g/cm3, with the target temperature of 310 K and the target pressure of 1.0 atm. The Berendsen thermostat method33 was used to control the system temperature. Finally, 100 ns MD simulation under the NVT ensemble was carried out via employing the periodic boundary condition. The SHAKE algorithm34 was applied to constrain all hydrogen-containing bonds with a tolerance

ACS Paragon Plus Environment

Journal of Chemical Information and Modeling

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

of 10−5 during the MD simulations. A cutoff of 12 Å was set for both van der Waals and electrostatic interactions. All the simulations were accomplished in AMBER12.32 To characterize the dynamics of ZA-loop during the ligand releasing process, the umbrella sampling technique 35 was employed to effectively capture all conformations along the ligand releasing pathway. The distance between MET107:Cand (+)-JQ1:C13 (see Figure S1) atoms was chosen as the reaction coordinate (RC). The MD simulations were performed with a series of biasing harmonic potential (10– 25 kcal/mol) along the RC from 13.0 Å to 30.0 Å (there were total about 56 simulation windows, and each window was separated by 0.3 Å). The equilibrium structure from the most nearby window was employed to be the initial structure for the next window. At least 20 ns MD simulations were carried out to check if it became stable for each window, via monitoring the RMSD values. And then additional 15 ns productive MD simulations were performed for sampling. Finally, the collected RC data (the last 10 ns MD trajectory for each window) were analyzed by the weighted histogram analysis method (WHAM) 36 to generate the potential of mean force (PMF) (namely free energy profile). Cluster analysis is a general unsupervised technique for finding patterns within data.37-39 In this work, the root mean-square deviation(RMSD)-based clustering was performed with ptraj, a simulation analysis tools implemented in AMBER12. As one of the most popular clustering algorithms, the average linkage cluster algorithm was used herein.40 Each cluster contained a representative structure whose RMSD was equidistant to all other cluster members. Structures were collected by sampling at 200 ps intervals, and cluster analysis was based on the heavy atoms of BRD4 protein to generate a sum of five clusters (only two most clusters were kept). The volumes of the substrate binding pocket were measured using POVME, an algorithm for measuring pocket volume.41 Structures were collected by sampling at 50 ps intervals and aligned to the initial crystal structure (PDB:3MXF). An inclusion sphere with a radius of 12 Å was defined that entirely encompassed the binding pocket. A written shell script was used to automatically calculate the volumes of all collected structures automatically. And all molecular figures shown in this article were created using PyMOL.42

ACS Paragon Plus Environment

Page 6 of 23

Page 7 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Chemical Information and Modeling

2. Born–Oppenheimer ab initio QM/MM MD Simulations. The equilibrium enzyme-ligand complex systems from the above MD simulations were cut into a sphere by removing the solvent water molecules beyond 30 Å of the C22 atom (as shown in Figure S1) of (+)-JQ1/(-)-JQ1. The resulting systems were then partitioned into QM and MM subsystems. The Asn140 and ligands were chosen as the QM subsystem (~ 65 atoms in total), which was treated by B3LYP

43, 44

with the 6-31G*

basis set. The QM/MM boundaries were described by the pseudobond approach with improved pseudobond parameters.45-48 All the remaining atoms were described by the same MM force field used in the previous classical MD simulations. For all QM/MM calculations, the spherical boundary condition 45, 48 was applied, and atoms more than 22 Å away from the center were fixed. Cutoffs of 18 and 12 Å were employed for the electrostatic and van der Waals interactions, respectively. There was no cutoff for the electrostatic interactions between QM and MM regions. After minimization of the prepared QM/MM systems, 25 ps QM/MM MD simulations were performed using a time step of 1 fs and the Beeman algorithm 49 to integrate the Newton equations of motion. The Berendsen thermostat method 33 was used to control the system temperature at 310 K. The configurations of the last 5 ps were collected for data analysis. All ab initio QM/MM calculations were performed in modified QChem-Tinker programs.50,

51

A similar QM/MM modeling protocol had been

extensively employed in our previous studies. 52-55

Figure 2. Time series of the skeleton-atoms RMSD for the two loops in BRD4.

ACS Paragon Plus Environment

Journal of Chemical Information and Modeling

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Results and discussion 1. Static ligand binding modes of JQ1 To probe the structural stability of the BRD4-JQ1 complex, the detailed RMSD evolution along the 100 ns MM MD simulations are summarized in Figure S2. It indicates that the BRD4 protein structures are very stable with backbone RMSD value near 2 Å after 40 ns MD simulations, and most fluctuation is owing to the loops. Further RMSD comparisons (see Figure 2) indicate that the BC-loop is as stable as observed in apo- BRD4 (see Figure S2). However, the ZA-loop is very flexible (from 0.5 to 2.5 Å) to achieve the final optimal binding modes of (+)-JQ1. Since the RMSD of ZA-loop is also around 2.5 Å in the stable apo- state (see Figure S2), we proposed that ligand-induced conformation change is unnecessary for the active inhibitor (+)-JQ1 binding to the active pocket (will be discussed again in section 2). Regarding to the inactive inhibitor (-)-JQ1, as its RMSD is near to 3 Å (see Figure 2.) which is bigger than that in the apo- BRD4, it indicates that ZA-loop motion is required to accommodate the ligand into the binding pocket. That is, the active pocket is extendible (will discuss it infra), which had also been proposed in previous studies 23, 26

. On the basis of the flat RMSD evolution for ZA-loop after about 40 ns simulations,

the inactive inhibitor (-)-JQ1 would be also well accommodated in BRD4, thus it is interesting to answer how the ZA-loop motion occur when (-)-JQ1 instead of (+)-JQ1 binding to active site? As shown in Figure 3, the structural superposition clearly reveals that BC-loop as well as the strong hydrogen bond with ligand are very stable either for (+)-JQ1 or (-)-JQ1. Moreover, the side group on the chiral center of (+)-JQ1 locates at the free space between two loops and towards the solvent outside the pocket, which is matching well to the original binding modes in XRD (x-ray diffraction) structure (PDB:3MXF), thus it shares the similar RMSD for ZA-loop in apo-BRD4 as mentioned above. In contrast, the ZA-loop is pushed out since the side group on the chiral center of (-)-JQ1 is orientated to the inside of the binding pocket and it would occupy the original location for L94 and L92 in ZA-loop, and thus brings bigger

ACS Paragon Plus Environment

Page 8 of 23

Page 9 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Chemical Information and Modeling

RMSD of ZA-loop as presented in Figure 2. As a result, the binding modes are different between the two racemates (see Figure 3.), the stable σ-π interaction is observed between L94 and a-ring of (+)-JQ1, while it is destroyed and replaced by another σ-π interaction between (-)-JQ1 and aromatic ring of Y97. Obviously, the Y97 will flip over from outside to inside pocket as the ZA-loop is pushed out by (-)-JQ1. In sum, the most significant difference is that the side chain on the chiral center is anchored by L94 and N140 for (+)-JQ1 whereas it is restricted by Y97 and N140 for (-)-JQ1 in the final optimal binding modes. As a distinguished ZA-loop motion is required upon (-)-JQ1 binding to BRD4, and then enlarge the binding pocket to accommodate it, further cluster analysis and pocket volumes calculation are performed to investigate the expansion of binding pocket. As shown in Figure 4, the volumes of active pocket is larger than 456 Å3 upon (-)-JQ1 binding to BRD4, while that is only about 298 Å3 for (+)-JQ1. Obviously, the pocket mouth is bigger for (-)-JQ1 than that for (+)-JQ1 binding into BRD4, which is consistent with the bigger distance between the two loops, as shown in Figure 3. Regarding to the apo-BRD4, the computationally predicted pocket volume is about 283.6±5.7 Å3, which is close to that for the (+)-JQ1 binding BRD4. It indicates that the active pocket opening is essential for binding with (-)-JQ1 whereas not necessary for (+)-JQ1. Foreseeably, much more water would enter into the active pocket if (-)-JQ1 binding into BRD4 with a bigger pocket mouth (see Figure S4.). As a result, the ligand binding environment will be totally different since the dielectric constant of the binding pocket would be modulated by the number of water molecules in the binding pocket, and leads to different binding affinity. Therefore, to reveal the functional roles of ZA-loop in binding JQ1, the state-of-the-art density functional QM/MM MD simulations were further performed to accurately analyze the protein-ligand interaction energy, as summarized in Figure 5 and S5. The total binding affinity (namely stabilization energy contributed from protein environment) is about -45 kcal/mol for (+)-JQ1, similar to that for (-)-JQ1 (-42 kcal/mol, see Figure 5a.), thus (-)-JQ1 would also have a high binding affinity in BRD4. Similarly, see from Figure 5b, the interaction energy between (-)-JQ1 and

ACS Paragon Plus Environment

Journal of Chemical Information and Modeling

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ZA-loop (-23 kcal/mol) is higher than that for (+)-JQ1 (-21 kcal/mol), without distinctive thermodynamics features between the two racemates. The only difference for the two racemates are the stabilization energy contributed from the L94 and Y97, as shown in Figure 5(c, d), L94 is a critical residue to stabilize the (+)-JQ1 (-4 kcal/mol) while Y97 is important in stabilizing the (-)-JQ1 (-5 kcal/mol), which is consistent with the different binding modes as observed in Figure 3. Experimentally, (-)-JQ1 had been proved to be inactive whereas (+)-JQ1 was active against BRD4 23. Herein, our modeling indicates that BRD4 could accommodate both of them as its could enlarge the binding pocket via the ZA-loop motion, which is consistent with the experimental results that several kinds of inhibitor with different skeleton are identified.19,

56

Indeed, the ZA-loop is very flexible to optimize the

binding modes of the inactive inhibitor (-)-JQ1 and makes it also thermodynamically favorable in comparison to the active inhibitor (+)-JQ1. In this case, it is experimentally and computationally incongruent when seen from “ligand-protein binding affinity viewpoint". Since the more flexible ZA-loop and bigger pocket volumes are observed for (-)-JQ1 in comparison to the (+)-JQ1 upon binding to BRD4, it raises up a hypothesis that the inactivity of (-)-JQ1 is due to the facility in releasing process or the difficulty in binding procedure in kinetics. (vide infra).

ACS Paragon Plus Environment

Page 10 of 23

Page 11 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Chemical Information and Modeling

Figure 3. The comparison of the ligand-binding-modes with key residues in ZA-loop (yellow one for (+)-JQ1 and green one for (-)-JQ1). Left is the structural overlap and the red double-head arrow indicates the distinguishable motion of ZA-loop, and the detailed distance evolution between the two loops (namely between Asn140:Cα and Leu94/Leu92:Cα) are shown in Figure S3. The two distinct σ−π interaction are highlighted in red in the right figures (the distance between carbon atom and the center of aromatic ring are noted).

Figure 4. Pocket surface illustration for two racemates of JQ1 binding to BRD4 based on cluster analysis. The model binding with (+)-JQ1 is left-rotated to make the pocket mouth directly face toward us for clear observation.

Figure 5. Interaction energy analysis between selected residues and ligand. (a) The total stabilization energy from protein environment for (+)-JQ1 and (-)-JQ1 ligands. (b) The interaction energy between ZA-loop and the two racemates. (c) The interaction energy between the L94 and

ACS Paragon Plus Environment

Journal of Chemical Information and Modeling

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

the two racemates. (d) The interaction energy between the Y97 and the two racemates. The detailed interaction energy analysis between ZA-loop/BC-loop and ligand are summarized in Figure S5. For the BC-loop, it presents similar interaction with the two ligands, especially, D144 is disadvantageous while K141 is advantageous in stabilizing the two racemates of JQ1. All interaction energies are calculated based on 5000 snapshots from the equilibrium QM/MM MD trajectories.

2. Binding/Releasing kinetics of JQ1 To investigate the dynamic releasing process of ligands, the free energy profiles for the (+)-JQ1 releasing from the active pocket are mapped out as shown in Figure 6. It indicates that the releasing of (+)-JQ1 is very facile kinetically, with less than 4 kcal/mol barrier, and involving three stages (see Figure 6 and 7). At the beginning of the first stage (see Figure 7a.), the strong hydrogen bond (HB) between N140 and (+)-JQ1 is maintained well, keeping ~ 2.6 Å between the two HB heavy atoms. And the above mentioned σ-π stacking interaction {Leu 94:CD1-- (+)-JQ1: a-ring}, as well as the σ-π stacking interaction between Ile146:CB and (+)-JQ1: d-ring, are very stable. At the end of the first stage (see Figure 7b.), both HB interaction with N140 and σ-π stacking interaction with Ile146 are kept well, while {Leu 94:CD1-- (+)-JQ1: a-ring} σ-π stacking interaction is destroyed and replaced by HB interaction mediated by a water molecule. During the second stage (see Figure 7c&d), the N140:ligand HB interaction and Ile146—ligand σ-π stacking interaction was broken. Instead, several water molecules entered into the pocket and surrounded the hydrophilic N140 and Ile146, as well as the polar group of (+)-JQ1, to constructing a new hydrogen bond network. At the end of the second stage, namely near to the highest barrier region as shown in Figure 6, since the σ-π stacking interactions with Leu92 and Leu94 were totally damaged, (+)-JQ1 is no longer anchored between ZA-loop and BC-loop, then occurred a molecular hopping and surround by several water finally. At the beginning of the third stage (see Figure 7e), (+)-JQ1 is almost got away from the active pocket, but still contacting with Lys99 (in ZA-loop) by hydrogen bond interaction, thus presents a flat region (namely local metastable state, point e) in the free energy profile as shown in Figure 6. When this HB is finally destroyed, the ligand was completely solvated in water solution at the end of the third stage (see Figure 7f).

ACS Paragon Plus Environment

Page 12 of 23

Page 13 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Chemical Information and Modeling

Regarding to the binding pocket, it would involve “closed”-“open”-“half-closed” three stages along the releasing process, which is consistent with the conformation change of ZA-loop, as shown in Figure 7. At the first stage (a&b), the secondary structure of ZA-loop is stable since the protein-ligand interaction is still very strong and maintained well. At the second stage, followed by the breakage of strong hydrogen bond between N140 and ligand, ZA-loop move forward to the outside solvent and enlarge the binding pocket as well as increase the water molecules in the active site. At the third stage, as ligand is finally released from the binding pocket, all of the interactions between ZA-loop and ligand are lost. From then on, the enzyme will recover to the apo- state spontaneously, thus ZA-loop move inwardly and shrink the binding pocket as well as many water molecules will be extruded. See from Figure S6, the water number evolution, as well as the distance evolution between secondary-structure-stable BC-loop and conformation-flexible ZA-loop along the ligand releasing, is highly consistent with the observed “closed”-“open”-“half-closed” kinetics. Since at the second stage, it is required to destroy very strong HB interaction with N140 as well as important σ-π stacking interaction with ZA-loop, which brings most hindering effect, thus it is the kinetically advantageous stage for preventing (+)-JQ1 releasing, to increase the so-called “drug-target” resident time. Intriguingly, a higher-energy-barrier “second stage” and unstable “third stage” are identified for (-)-JQ1, as shown in Figure 8. Therefore, the releasing of (-)-JQ1 is not as facile as (+)-JQ1 releasing from the pocket. For the binding kinetics procedures, considering the bigger active pocket is essential to nestle the (-)-JQ1 as observed in Figure 4, the ligand-induced ZA-loop conformation change is required to open the pocket as we mention in section 1. Our further steered molecular dynamics (SMD) and simulated annealing(SA)/productive MD simulations (see detailed computational protocol in Figure S7) observed the binding kinetics difference between (+)-JQ1 and (-)-JQ1. Totally 20 independent MD trajectories for (+)-JQ1 and (-)-JQ1 are obtained respectively, and the distance between JQ1 and Asn140 which represents the strong hydrogen bond between BRD4 and JQ1 (highlight in red dash-line in Figure 3 and S1) is selected to detect whether

ACS Paragon Plus Environment

Journal of Chemical Information and Modeling

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(+)-JQ1/(-)-JQ1 ligands binding to the active site. It is found that (+)-JQ1 would successfully enter into the binding sites four times during the simulated annealing process spontaneously and would not immigrate backward into the solvent any more, whereas (-)-JQ1 could not enter into the binding sites even if the ligand had been steered to the mouth of the active site, as shown in Figure 9 (see more details in Figure S8). It reveals that (+)-JQ1 is facile to binding into the pocket, which leads to its achievable binding kinetics and is consistent with its well inhibitory effect (~77 nM). In contrast, the (-)-JQ1 is difficult to enter into the binding site and leads to a worse inhibitory effect as experiments observed (>10,000 nM). To figure out the key regulatory factors on the binding kinetics difference between (+)-JQ1 and (-)-JQ1, further structural analysis are summarized in Figure 10. The weak interaction between (+)-JQ1 and Y94 is not observed in the 30 ns SA and productive MD simulation (See Figure 10a). However, the very stable hydrogen bond with the conserved Asn140 residue is formed spontaneously and maintained well (see Figure 10a), which is as strong as observed in XRD structure and the above mentioned MD simulations as shown in Figure 3, which share a highly similar binding modes (see Figure 10b). In contrast, the significant important conserved hydrogen bond interaction between ligand and Asn140 is totally lost. Meanwhile, instead of the π-stacking interaction with Y97, (-)-JQ1 would maintain very stable π-stacking with P82 and locate close to the ZA-loop. The (-)-JQ1 could partially enter into the active pocket to keep at a metastable binding state (see Figure 10c) in the enlarged pocket due to ZA-loop motion. Nevertheless, the binding modes (see Figure 10d) are indeed different from the above mentioned thermodynamically favorable state as shown in Figure 3. And the escapement of (-)-JQ1 from active pocket mouth to the solvent could be observed in several MD trajectories (take one example in Figure S8). That is, it is kinetically unfavorable to achieve the more stable binding modes due to lack of conserved hydrogen bond interaction with Asn140 to further pull the ligand entering into the bottom of the active pocket site. Therefore, we conclude that Asn140 is not only critical factor in “ligand-protein binding affinity” but also in facilitating control along the “inhibitor binding kinetics”.

ACS Paragon Plus Environment

Page 14 of 23

Page 15 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Chemical Information and Modeling

Figure 6. The free energy profiles of (+)-JQ1 releasing along the defined reaction coordinate. The representative structures at the selected points (a/b/c/d/e/f) are shown in Figure 7. The benchmark test on the convergence of PMF curve are provided in Figure S9 and S10.

Figure 7. The representative structures in the release process of (+)-JQ1 from BRD4. (a,b) represent the first stage, (c,d) show the second stage and (e,f) reveal the third stage. The detailed location of (a-f) along the release process is referred to Figure 6. The residues playing key roles in ligand releasing are highlighted in purple and most water molecules in the pocket are shown in red ball. The key hydrogen bonds are shown in red-dashed line and π-stacking interaction is noted by blue-dashed line. BC-loop is colored in blue and ZA-loop is colored in wheat.

ACS Paragon Plus Environment

Journal of Chemical Information and Modeling

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 8. The free energy profiles of (-)-JQ1 releasing along the defined reaction coordinate. The Standard Deviation (0.1 kcal/mol) is estimated with the same methods shown in Figure S9.

Figure 9. The representative distance evolution between Asn140:N and JQ1:N (the strong hydrogen bond (HB) heavy-atom between N140 and JQ1, see Figure S1.). Another 3 representative spontaneously binding trajectories for (+)-JQ1 and (-)-JQ1 are summarized in Figure S8. The (-)-JQ1 would form hydrogen bond with Asn140 during the simulated annealing(SA) process (~2.5-5 ns in this case). However, it would lost at the end of the SA procedure (~5-8 ns) and would not reconstruct the hydrogen bond interaction during the productive MD simulations (8ns later). Moreover, (-)-JQ1 might move backward into the solvent again as observed in some case as presented in Figure S8.

ACS Paragon Plus Environment

Page 16 of 23

Page 17 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Chemical Information and Modeling

Figure 10. The representative binding modes of (+)-JQ1 and (-)-JQ1 (yellow one for (+)-JQ1 (a) and green one for (-)-JQ1 (c)) at the equilibrium state from the spontaneously binding trajectories (refer to Figure 9), and the corresponding structural superposition referring to the optimal binding models (in gray) mentioned in Figure 3 ((b) for (+)JQ1 and (d) for (-)-JQ1 respectively).

Conclusion By atomic-level MD simulations, the accessibility of the active pocket ranges from fully closed to completely opening state regulating by ZA-loop in BRD4 is investigated directly. And by further electronic-level Born–Oppenheimer density functional QM/MM MD simulations, the optimal binding modes of inhibitor (+)-JQ1 are identified, and critical roles of ZA-loop in stabilizing the ligands are also clarified. It is found that not only the well-known conserved residue Asn140 in BC-loop but also the newly characterized L92/L94/Y97 residues in ZA-loop are playing important functional roles on (+)-JQ1 binding and releasing dynamics. Furthermore, in comparison with the inactive inhibitor (-)-JQ1, (+)-JQ1 is proved to be both thermodynamically reasonable and kinetics achievable towards BRD4, which is

ACS Paragon Plus Environment

Journal of Chemical Information and Modeling

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

consistent with the high inhibitory activity for (+)-JQ1 as observed in previous experiments.23 In contrast, the failure of (-)-JQ1 to be an effective inhibitor is due to the unfeasible binding kinetics features against BRD4. It is found that Asn140 is not only important in “ligand-protein binding affinity” but also critical in “binding kinetics” for BRD4 inhibition. This work emphasize that binding kinetics characteristics (both include binding and releasing) should be carefully considered even more important than the thermodynamically binding affinity for future selective inhibitor design towards BRD.57 As much as we known, the “drug-target” resident time has been emphasized to promote the drug efficacy recently.56-58 On the basis of our simulations, (+)-JQ1 is easy to binding but also very facile to releasing from BRD4, thus, even if (+)-JQ1 showed high inhibitory effect experimentally and the strong binding affinity with BRD4 is confirmed herein computationally, further structural modification to hinder the releasing of (+)-JQ1 would be very helpful to increase the “drug-target” resident-time, as well as guidance for accelerating the other domain-selective inhibitor design among BET family.

ASSOCIATED CONTENT Supporting Information Figure S1-S10. This material is available free of charge via the Internet at http://pubs.acs.org.

AUTHOR INFORMATION Corresponding Authors: *E-mail: [email protected] *E-mail: [email protected]

Notes The authors declare no competing financial interest.

ACS Paragon Plus Environment

Page 18 of 23

Page 19 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Chemical Information and Modeling

Acknowledgments This work was supported by the National Science Foundation of China (21203257, 21272289) and Pearl River S&T Nova Program of Guangzhou (2014J2200062). We sincerely acknowledge the funding granted from Department of Education of Guangdong Province supporting national major cultivation project for integrated medicine against fatty liver disease. We also thank the National Supercomputing Centers in Shenzhen and Guangzhou for providing the computational resources.

References (1) Kouzarides, T., Chromatin Modifications and Their Function. Cell 2007, 128, 693-705. (2) Arrowsmith, C. H.; Bountra, C.; Fish, P. V.; Lee, K.; Schapira, M., Epigenetic Protein Families: A New Frontier for Drug Discovery. Nat Rev Drug Discov. 2012, 11, 384-400. (3) Heightman, T. D., Therapeutic Prospects for Epigenetic Modulation. Expert Opin. Ther. Tar. 2011, 15, 729-740. (4) Denis, G. V., Bromodomain Coactivators in Cancer, Obesity, Type 2 Diabetes, and Inflammation. Discov. Med. 2010, 55, 489-499. (5) Sanchez, R.; Zhou, M. M., The Role of Human Bromodomains in Chromatin Biology and Gene Transcription. Curr. Opin. Drug. Disc.2009, 12, 659-665. (6) Bannister, A. J.; Kouzarides, T., Regulation of Chromatin by Histone Modifications. Cell Res. 2011, 21, 381-395. (7) Rando, O. J., Combinatorial Complexity in Chromatin Structure and Function: Revisiting The Histone Code. Curr. Opin. Genet. Dev. 2012, 22, 148-155. (8) Turner, B. M., Reading Signals on The Nucleosome with A New Nomenclature for Modified Histones. Nat. Struct. Mol. Biol. 2005, 12, 110-112. (9) Jenuwein, T.; Allis, C. D., Translating The Histone Code. Science 2001, 293, 1074-1080. (10) Lorenzen, J. M.; Martino, F.; Thum, T., Epigenetic Modifications In Cardiovascular Disease. Basic Res. Cardiol. 2012, 107,1-10. (11) Arrowsmith, C. H.; Bountra, C.; Fish, P. V.; Lee, K.; Schapira, M., Epigenetic Protein Families: A New Frontier for Drug Discovery. Nat. Rev. Drug. Discov. 2012, 11, 384-400. (12) Dawson, M. A.; Kouzarides, T., Cancer Epigenetics: From Mechanism to Therapy. Cell 2012, 150, 12-27. (13) Chi, P.; Allis, C. D.; Wang, G. G., Covalent Histone Modifications - Miswritten, Misinterpreted and Mis-Erased in Human Cancers. Nat. Rev. Cancer 2010, 10, 457-469. (14) Dhalluin, C.; Carlson, J. E.; Zeng, L.; He, C.; Aggarwal, A. K.; Zhou, M. M., Structure and Ligand of A Histone Acetyltransferase Bromodomain. Nature 1999, 399, 491-496. (15) Furdas, S. D.; Carlino, L.; Sippl, W.; Jung, M., Inhibition of Bromodomain-Mediated Protein-Protein Interactions as A Novel Therapeutic Strategy. Medchemcomm 2012, 3, 123-134.

ACS Paragon Plus Environment

Journal of Chemical Information and Modeling

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(16) Filippakopoulos, P.; Picaud, S.; Mangos, M.; Keates, T.; Lambert, J. P.; Barsyte-Lovejoy, D.; Felletar, I.; Volkmer, R.; Muller, S.; Pawson, T.; Gingras, A. C.; Arrowsmith, C. H.; Knapp, S., Histone Recognition and Large-Scale Structural Analysis of The Human Bromodomain Family. Cell 2012, 149, 214-231. (17) Muvva, C.; Singam, E. R. A.; Raman, S. S.; Subramanian, V., Structure-Based Virtual Screening of Novel, High-Affinity BRD4 Inhibitors. Mol. Biosyst. 2014, 10, 2384-2397. (18) Zhao, H. T.; Gartenmann, L.; Dong, J.; Spiliotopoulos, D.; Caflisch, A., Discovery of BRD4 Bromodomain Inhibitors by Fragment-Based High-Throughput Docking. Bioorg. Med. Chem. Lett. 2014, 24, 2493-2496. (19) Hewings, D. S.; Rooney, T. P. C.; Jennings, L. E.; Hay, D. A.; Schofield, C. J.; Brennan, P. E.; Knapp, S.; Conway, S. J., Progress in The Development and Application of Small Molecule Inhibitors of Bromodomain-Acetyl-Lysine Interactions. J. Med. Chem. 2012, 55, 9393-9413. (20) Muller, S.; Filippakopoulos, P.; Knapp, S., Bromodomains as Therapeutic Targets. Expert Rev. Mol. Med. 2011, 13, 1-21. (21) Dawson, M. A.; Prinjha, R. K.; Dittmann, A.; Giotopoulos, G.; Bantscheff, M.; Chan, W. I.; Robson, S. C.; Chung, C. W.; Hopf, C.; Savitski, M. M.; Huthmacher, C.; Gudgin, E.; Lugo, D.; Beinke, S.; Chapman, T. D.; Roberts, E. J.; Soden, P. E.; Auger, K. R.; Mirguet, O.; Doehner, K.; Delwel, R.; Burnett, A. K.; Jeffrey, P.; Drewes, G.; Lee, K.; Huntly, B. J. P.; Kouzarides, T., Inhibition of Bet Recruitment to Chromatin as An Effective Treatment for Mll-Fusion Leukaemia. Nature 2011, 478, 529-533. (22) Delmore, J. E.; Issa, G. C.; Lemieux, M. E.; Rahl, P. B.; Shi, J. W.; Jacobs, H. M.; Kastritis, E.; Gilpatrick, T.; Paranal, R. M.; Qi, J.; Chesi, M.; Schinzel, A. C.; Mckeown, M. R.; Heffernan, T. P.; Vakoc, C. R.; Bergsagel, P. L.; Ghobrial, I. M.; Richardson, P. G.; Young, R. A.; Hahn, W. C.; Anderson, K. C.; Kung, A. L.; Bradner, J. E.; Mitsiades, C. S., BET Bromodomain Inhibition as A Therapeutic Strategy to Target C-Myc. Cell 2011, 146, 903-916. (23) Filippakopoulos, P.; Qi, J.; Picaud, S.; Shen, Y.; Smith, W. B.; Fedorov, O.; Morse, E. M.; Keates, T.; Hickman, T. T.; Felletar, I.; Philpott, M.; Munro, S.; Mckeown, M. R.; Wang, Y.; Christie, A. L.; West, N.; Cameron, M. J.; Schwartz, B.; Heightman, T. D.; La Thangue, N.; French, C. A.; Wiest, O.; Kung, A. L.; Knapp, S.; Bradner, J. E., Selective Inhibition of BET Bromodomains. Nature 2010, 468, 1067-1073. (24) Mertz, J. A.; Conery, A. R.; Bryant, B. M.; Sandy, P.; Balasubramanian, S.; Mele, D. A.; Bergeron, L.; Sims, R. J., Targeting Myc Dependence in Cancer by Inhibiting BET Bromodomains. Proc. Natl. Acad. Sci. U.S.A. 2011, 108, 16669-16674. (25) Nicodeme, E.; Jeffrey, K. L.; Schaefer, U.; Beinke, S.; Dewell, S.; Chung, C. W.; Chandwani, R.; Marazzi, I.; Wilson, P.; Coste, H.; White, J.; Kirilovsky, J.; Rice, C. M.; Lora, J. M.; Prinjha, R. K.; Lee, K.; Tarakhovsky, A., Suppression of Inflammation by A Synthetic Histone Mimic. Nature 2010, 468, 1119-1123. (26) Steiner, S.; Magno, A.; Huang, D.; Caflisch, A., Does Bromodomain Flexibility Influence Histone Recognition? Febs Lett. 2013, 587, 2158-63. (27) Hornak, V.; Abel, R.; Okur, A.; Strockbine, B.; Roitberg, A.; Simmerling, C., Comparison of Multiple Amber Force Fields and Development of Improved Protein Backbone Parameters. Proteins: Struct., Funct., Bioinf. 2006, 65, 712-725. (28) J. Wang, R. M. W., J.W. Caldwell, P.A. Kollman, D.A. Case, Development and Testing of A General Amber Force Field. J. Comput. Chem. 2004, 25, 1157-1174.

ACS Paragon Plus Environment

Page 20 of 23

Page 21 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Chemical Information and Modeling

(29) C.I. Bayly, P. C., W. Cornell, P.A. Kollman, A Well-Behaved Electrostatic Potential Based Method Using Charge Restraints for Deriving Atomic Charges The Resp Model. J. Phys. Chem. 1993, 97, 10269-10280. (30) Frisch, M. J. T., G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H. P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, J. A., Jr.; Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin, K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.; Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega, N.; Millam, J. M.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.; Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas, O.; Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian 09, Gaussian. Inc: Wallingford CT, 2009. (31) Jorgensen, W. L.; Chandrasekhar, J.; Madura, J. D.; Impey, R. W.; Klein, M. L., Comparison of Simple Potential Functions for Simulating Liquid Water. J. Chem. Phys. 1983, 79, 926-935. (32) Case, D. A.; Cheatham, T. E.; Darden, T.; Gohlke, H.; Luo, R.; Merz, K. M.; Onufriev, A.; Simmerling, C.; Wang, B.; Woods, R. J., The Amber Biomolecular Simulation Programs. J. Comput. Chem. 2005, 26, 1668-1688. (33) Berendsen, H. J. C.; Postma, J. P. M.; Van Gunsteren, W. F.; Dinola, A.; Haak, J. R., Molecular Dynamics with Coupling to An External Bath. J. Chem. Phys. 1984, 81, 3684. (34) Ryckaert, J.-P.; Ciccotti, G.; Berendsen, H. J. C., Numerical Integration of The Cartesian Equations of Motion of A System with Constraints: Molecular Dynamics of N-Alkanes. J. Comput. Phys. 1977, 23, 327-341. (35) Vijayaraj, R.; Van Damme, S.; Bultinck, P.; Subramanian, V., Molecular Dynamics and Umbrella Sampling Study of Stabilizing Factors in Cyclic Peptide-Based Nanotubes. J. Phys. Chem. B. 2012, 116, 9922-9933. (36) Grossfield, A., Wham: An Implementation of The Weighted Histogram Analysis Method, Http://Membrane.Urmc.Rochester.Edu/Content/Wham/, Version2.0.6ed. (accessed June 15, 2014). (37) Estacio, S. G.; Moreira, R.; Guedes, R. C., Characterizing The Dynamics and Ligand-Specific Interactions in The Human Leukocyte Elastase Through Molecular Dynamics Simulations. J. Chem. Inf. Model. 2011, 51, 1690-1702. (38) Amaro, R. E.; Swift, R. V.; Votapka, L.; Li, W. W.; Walker, R. C.; Bush, R. M., Mechanism of 150-Cavity Formation in Influenza Neuraminidase. Nat. Commun. 2011, 2,1-7. (39) Colombo, G.; Morra, G.; Meli, M.; Verkhivker, G., Understanding Ligand-Based Modulation of The Hsp90 Molecular Chaperone Dynamics at Atomic Resolution. Proc. Natl. Acad. Sci. U.S.A. 2008, 105, 7976-7981. (40) Shao, J. Y.; Tanner, S. W.; Thompson, N.; Cheatham, T. E., Clustering Molecular Dynamics Trajectories: 1. Characterizing The Performance of Different Clustering Algorithms. J. Chem. Theory Comput. 2007, 3, 2312-2334. (41) Durrant, J. D.; De Oliveira, C. A. F.; Mccammon, J. A., Povme: An Algorithm for Measuring Binding-Pocket Volumes. J. Mol. Graph. Model. 2011, 29, 773-776. (42) The Pymol Molecular Graphics System. Version 1.6; Schrödinger Llc: New York, 2013. (43) Lee, C.; Yang, W.; Parr, R. G., Development of The Colle-Salvetti Correlation-Energy Formula

ACS Paragon Plus Environment

Journal of Chemical Information and Modeling

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Into A Functional of The Electron Density. Phys. Rev. B 1988, 37, 785-789. (44) Becke, A. D., Density-Functional Exchange-Energy Approximation with Correct Asymptotic Behavior. Phys. Rev. A 1988, 38, 3098-3100. (45) Zhang, Y. K., Pseudobond Ab Initio Qm/Mm Approach and Its Applications to Enzyme Reactions. Theor. Chem. Acc. 2006, 116, 43-50. (46) Zhang, Y. K., Improved Pseudobonds for Combined Ab Initio Quantum Mechanical/Molecular Mechanical Methods. J. Chem. Phys. 2005, 122, 024114 (1-7). (47) Zhang, Y. K.; Liu, H. Y.; Yang, W. T., Free Energy Calculation on Enzyme Reactions with An Efficient Iterative Procedure to Determine Minimum Energy Paths on A Combined Ab Initio Qm/Mm Potential Energy Surface. J. Chem. Phys. 2000, 112, 3483-3492. (48) Zhang, Y. K.; Lee, T. S.; Yang, W. T., A Pseudobond Approach to Combining Quantum Mechanical and Molecular Mechanical Methods. J. Chem. Phys. 1999, 110, 46-54. (49) Beeman, D., Some Multistep Methods for Use In Molecular Dynamics Calculations. J. Comput. Phys. 1976, 20, 130-139. (50) Shao, Y. M., L. F.; Jung, Y.; Kussmann, J. O. C.; Brown, S. T.; Gilbert, A. T.; Slipchenko, L. V.; Levchenko, S. V.; O'neill, D. P.; Distasio, R. A.; Lochan, R. C.; Wang, T.; Beran, G. J.; Besley, N. A.; Herbert, J. M.; Lin, C. Y.; Van Voorhis, T.; Chien, S. H.; Sodt, A.; Steele, R. P.; Rassolov, V. A.; Maslen, P. E.; Korambath, P. P.; Adamson, R. D.; Austin, B.; Baker, J.; Byrd, E. F.; Dachsel, H.; Doerksen, R. J.; Dreuw, A.; Dunietz, B. D.; Dutoi, A. D.; Furlani, T. R.; Gwaltney, S. R.; Heyden, A.; Hirata, S.; Hsu, C. P.; Kedziora, G.; Khalliulin, R. Z.; Klunzinger, P.; Lee, A. M.; Lee, M. S.; Liang, W.; Lotan, I.; Nair, N.; Peters, B.; Proynov, E. I.; Pieniazek, P. A.; Rhee, Y. M.; Ritchie, J.; Rosta, E.; Sherrill, C. D.; Simmonett, A. C.; Subotnik, J. E.; Woodcock, H. L.; Zhang, W.; Bell, A. T.; Chakraborty, A. K.; Chipman, D. M.; Keil, F. J.; Warshel, A.; Hehre, W. J.; Schaefer, H. F.; Kong, J.; Krylov, A. I.; Gill, P. M.; Head-Gordon, M. Q-Chem,Version 3.0,Q-chem,Inc.:Pittsburgh,PA. 2006. (51) Ponder, J. W.TINKER,Software Tools for Molecular Design,Version 4.2. 2004. (52) Wu, R. B.; Gong, W. J.; Ting, L.; Zhang, Y. K.; Cao, Z. X., Qm/Mm Molecular Dynamics Study of Purine-Specific Nucleoside Hydrolase. J. Phys. Chem. B. 2012, 116, 1984-1991. (53) Wu, R. B.; Lu, Z. Y.; Cao, Z. X.; Zhang, Y. K., Zinc Chelation with Hydroxamate in Histone Deacetylases Modulated by Water Access to The Linker Binding Channel. J. Am. Chem. Soc. 2011, 133, 6110-6113. (54) Zhou, J. W.; Xie, H. J.; Liu, Z. H.; Luo, H. B.; Wu, R. B., Structure-Function Analysis of The Conserved Tyrosine and Diverse Pi-Stacking Among Class I Histone Deacetylases: A Qm (Dft)/Mm Md Study. J. Chem. Inf. Model. 2014, 54, 3162-3171. (55) Chen, N. H.; Zhou, J. W.; Li, J. B.; Xu, J.; Wu, R. B., Concerted Cyclization of Lanosterol C-Ring and D-Ring Under Human Oxidosqualene Cyclase Catalysis: An Ab Initio Qm/Mm Md Study. J. Chem. Theory Comput. 2014, 10, 1109-1120. (56) Brand, M.; Measures, A. M.; Wilson, B. G.; Cortopassi, W. A.; Alexander, R.; Hoss, M.; Hewings, D. S.; Rooney, T. P. C.; Paton, R. S.; Conway, S. J., Small Molecule Inhibitors of Bromodomain-Acetyl-Lysine Interactions. Acs Chem. Biol. 2015, 10, 22-39. (57) Picaud, S.; Wells, C.; Felletar, I.; Brotherton, D.; Martin, S.; Savitsky, P.; Diez-Dacal, B.; Philpott, M.; Bountra, C.; Lingard, H.; Fedorov, O.; Muller, S.; Brennan, P. E.; Knapp, S.; Filippakopoulos, P., Rvx-208, An Inhibitor of BET Transcriptional Regulators with Selectivity for The Second Bromodomain. Proc. Natl. Acad. Sci. U.S.A. 2013, 110, 19754-19759. (58) Zhang, R.; Monsma, F. Binding Kinetics and Mechanism of Action: Toward The Discovery and

ACS Paragon Plus Environment

Page 22 of 23

Page 23 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Chemical Information and Modeling

Development of Better and Best in Class Drugs. Expert. Opin. Drug. Discov. 2010, 5, 1023-1029. (59) Bai, F.; Xu, Y.; Chen, J.; Liu, Q.; Gu, J.; Wang, X.; Ma, J.; Li, H.; Onuchic, J. N.; Jiang, H. Free Energy Landscape for The Binding Process of Huperzine A to Acetylcholinesterase. Proc. Natl. Acad. Sci. USA. 2013, 110, 4273-4278. (60) Kral, A. M.; Ozerova, N.; Close, J.; Jung, J.; Chenard, M.; Fleming, J.; Haines, B. B.; Harrington, P.; Maclean, J.; Miller, T. A.; Secrist, P.; Wang, H.; Heidebrecht, R. J. Divergent Kinetics Differentiate The Mechanism of Action of Two Hdac Inhibitors. Biochemistry. 2014, 53, 725-734.

ACS Paragon Plus Environment